Correlation between familial cancer history and epidermal growth factor receptor mutations in Taiwanese never smokers with non-small cell lung cancer: a case-control study

Abstract

Background

Lung cancer is a leading cause of cancer deaths in the world. Cigarette smoking remains a prominent risk factor, but lung cancer incidence has been increasing in never smokers. Genetic abnormalities including epidermal growth factor receptor (EGFR) mutations predominate in never smoking lung cancer patients. Furthermore, familial aggregations of patients with these mutations reflect heritable susceptibility to lung cancer. The correlation between familial cancer history and EGFR mutations in never smokers with lung cancer requires investigation.

Methods

This was a retrospective case-control study that evaluated the prevalence of EGFR mutations in lung cancer patients with familial cancer history. Never smokers with lung cancer treated at a hospital in Taiwan between April 2012 and May 2014 were evaluated. Inclusion criteria were never smokers with non-small cell lung cancer (NSCLC). Exclusion criteria involved patients without records of familial cancer history or tumor genotype.

Results

This study included 246 never smokers with lung cancer. The study population mainly involved never smoking women with a mean age of 60 years, and the predominant tumor histology was adenocarcinoma. Lung cancer patients with familial cancer history had an increased prevalence of EGFR mutations compared to patients without family history [odds ratio (OR): 5.9; 95% confidence interval (CI): 3.3-10.6; P<0.001]. Specifically, 57 out of 85 cancer patients (67%) with familial cancer history had these mutations, while 41 out of 161 patients (25%) without family history harbored mutations. Subgroup analysis also revealed that patients with familial lung cancer history had stronger association with EGFR mutations (OR: 7.5; 95% CI: 3.4-16.3; P<0.001) compared to patients with family history of non-pulmonary cancers (OR: 5.0; 95% CI: 2.5-10.0; P<0.001).

Conclusions

The study demonstrated an increased prevalence of EGFR mutations in Taiwanese never smoking lung cancer patients with familial cancer history. Moreover, a sizable proportion of never smoking cancer patients harbored these mutations. These observations have implications for the treatment of lung cancer in never smokers.

Introduction

Lung cancer is a leading cause of cancer deaths worldwide with more than one million deaths each year (1). Patients often present with advanced disease at diagnosis and are ineligible for curative treatment (2). Despite advances in therapy, prognosis has been poor with 5-year survival rates of 16% for affected patients (3). Cigarette smoking remains a prominent risk factor for developing lung cancer and accounts for the majority of cancer deaths (4). Consequently, previous research has focused on lung cancer in cigarette smokers.

However, lung cancer incidence has been increasing in never smokers (5). Never smokers now constitute nearly 25% of lung cancer patients worldwide, a trend that is especially prominent in Asia, where 50% of women with lung cancer never smoke (6,7). Risk factors other than tobacco contribute to cancer development, including occupational carcinogens, air pollution, cooking fumes, and genetic abnormalities (8). Previous studies have linked specific genetic abnormalities to lung cancer development (9-11).

Furthermore, lung cancer patients often reported first degree relatives with cancer, reflecting potential genetic factors associated with carcinogenesis (12). Although definite evidence of heritable genetic mutations remains elusive, familial aggregations of lung cancer patients with epidermal growth factor receptor (EGFR) mutations have been documented (13). A previous investigation of EGFR mutations and familial cancer history in lung cancer patients reported an insignificant association, but the cohort mainly involved Caucasians who were less susceptible to EGFR mutations (14). Since Asians and Caucasians with lung cancer have different prevalence of EGFR mutations, the relationship between familial cancer history and EGFR mutations in these ethnic groups may differ (15). This study aims to clarify the correlation between familial cancer history and the prevalence of EGFR mutations in Taiwanese never smokers with lung cancer.

Patients and methods

Patient selection

Never smokers with lung cancer treated at a tertiary hospital in Taiwan from April 2012 to May 2014 were evaluated for eligibility. Inclusion criteria were never smokers, defined by a lifetime use of less than 100 cigarettes, with a diagnosis of non-small cell lung cancer (NSCLC). Patients must have records of tumor genotype and familial cancer history for inclusion. Tumor samples were obtained by bronchoscopic or image guided biopsy and subsequently genotyped using direct polymerase chain reaction-based EGFR mutation tests. Exclusion criteria involved patients without records of familial cancer history or tumor genotype. The authors’ institution does not require review of retrospective studies involving de-identified medical records.

Data collection

Data including age, gender, disease stage, histology, tumor genotype, and familial cancer history were collected from medical records. Lung cancer patients with familial cancer history were allocated to cases, while lung cancer patients without family history were designated as controls. Familial cancer history was positive if any first or second degree relative had a diagnosis of cancer. However, due to the retrospective nature of family history collection, further sub-classification of familial heritage was not feasible. The study outcome was the prevalence of EGFR mutations in lung cancer patients with familial cancer history compared to those without family history. Patients with familial cancer history were further divided into those with family history of lung cancer and those with family history of non-pulmonary cancers. Finally, in the subset of patients with familial cancer history, the correlation between EGFR mutations and selected clinical characteristics was investigated.

Statistical analysis

Demographic characteristics of cases and controls were compared using Pearson’s chi-squared test for categorical variables and t-test for continuous variables. Pearson’s chi-squared test extrapolated the correlation between familial cancer history and EGFR mutations in lung cancer patients. Cross tabulation enabled calculations of odds ratio (OR) and 95% confidence interval (CI). All P values were based on a two-sided hypothesis. Analyses were performed using International Business Machines Statistical Package for the Social Sciences version 20 (SPSS Inc. Chicago, USA).

Results

Description of subjects

This study assessed 246 never smokers with lung cancer for eligibility. All patients had records of tumor genotype and familial cancer history, and therefore no exclusion of subjects occurred during analysis.

Demographic characteristics of cases and controls are shown in Table 1. Baseline characteristics including age, gender, disease stage, and histology showed no statistically significant difference between groups. The study population mainly involved never smoking women with a mean age of 60 years, and the predominant tumor histology was adenocarcinoma. The overall prevalence of EGFR mutations was 98 out of 246 patients (40%).

Table 1. Demographic characteristics of the study population.

Characteristics	Familial cancer history, n [%]	No familial cancer history, n [%]	Total, n [%]	P value
Total	85	161	246	0.89
Gender
Female	76 [89]	143 [89]	219 [89]
Male	9 [11]	18 [11]	27 [11]
Age, mean ± SD (years)	59±12	60±11	60±12	0.86
Histology				0.15
Adenocarcinoma	82 [97]	146 [91]	228 [93]
Squamous cell carcinoma	1 [1]	12 [7]	13 [5]
Others/mixed	2 [2]	3 [2]	5 [2]
EGFR mutation				<0.001
Negative	28 [33]	120 [75]	148 [60]
Positive	57 [67]	41 [25]	98 [40]
Cancer stage				0.16
Early (I-III)	27 [32]	66 [41]	93 [38]
Advanced (IV)	58 [68]	95 [59]	153 [62]

P value, Pearson’s chi-squared test for categorical variables and t-test for continuous variables; SD, standard deviation; n, number of patients; EGFR, epidermal growth factor receptor.

Familial cancer history and EGFR mutations

Lung cancer patients with familial cancer history had an increased prevalence of EGFR mutations compared to patients without family history (OR: 5.9; 95% CI: 3.3-10.6; P<0.001). Specifically, 57 out of 85 lung cancer patients (67%) with familial cancer history harbored EGFR mutations, while 41 out of 161 cancer patients (25%) without family history had these mutations. Table 2 summarizes these findings.

Table 2. Correlation between familial cancer history and EGFR mutations in never smokers with NSCLC.

Characteristics	Negative EGFR mutation, n [%]	Positive EGFR mutation, n [%]	Total, n	OR (P value)	95% CI
Family history of any cancer				5.9 (<0.001)	3.3-10.6
Yes	28 [33]	57 [67]	85
No	120 [75]	41 [25]	161
Family history of lung cancer				7.5 (<0.001)	3.4-16.3
Yes	11 [28]	28 [72]	39
No	120 [75]	41 [25]	161
Family history of non-pulmonary cancer				5.0 (<0.001)	2.5-10.0
Yes	17 [37]	29 [63]	46
No	120 [75]	41 [25]	161

EGFR, epidermal growth factor receptor; NSCLC, non-small cell lung cancer; OR, odds ratio; CI, confidence interval; n, number of patients; No, no family history of cancer.

Family history of lung cancer and EGFR mutations

Patients reporting family history specifically of lung cancer had higher prevalence of EGFR mutations compared to patients without familial cancer history (OR: 7.5; 95% CI: 3.4-16.3; P<0.001). Specifically, 28 out of 39 cancer patients (72%) with family history of lung cancer carried EGFR mutations, while 41 out of 161 patients (25%) without familial cancer history harbored these mutations. Importantly, there was a stronger association between EGFR mutations and familial lung cancer history compared to family history of any cancer. These observations are presented in Table 2.

Family history of non-pulmonary cancers and EGFR mutations

Lung cancer patients with family history of non-pulmonary cancers had an increased prevalence of EGFR mutations compared to patients without family history (OR: 5.0; 95% CI: 2.5-10.0; P<0.001). Specifically, 29 out of 46 cancer patients (63%) with family history of non-pulmonary cancers harbored EGFR mutations, whereas 41 out of 161 patients (25%) without any familial cancer history carried these mutations. The findings are presented in Table 2.

Familial cancer history and clinical characteristics

In the subset of patients with familial cancer history, those with EGFR mutations were associated with advanced stage IV disease (OR: 2.1; 95% CI: 0.8-5.4; P=0.12), although the trend did not achieve statistical significance. Lung cancer patients with EGFR mutations were more frequently female (OR: 4.9; 95% CI: 1.1-21.4; P=0.03), while no difference in age was observed regardless of EGFR mutation status. These findings are presented in Table 3.

Table 3. Correlation between EGFR mutations and clinical characteristics in never smoking lung cancer patients with familial cancer history.

Characteristics	Negative EGFR mutation, n [%]	Positive EGFR mutation, n [%]	Total, n	OR (P value)	95% CI
Lung cancer stage				2.1 (0.12)	0.8-5.4
Early (I-III)	12 [44]	15 [56]	27
Advanced (IV)	16 [27]	42 [73]	58
Gender				4.9 (0.03)	1.1-21.4
Female	22 [29]	54 [71]	76
Male	6 [67]	3 [33]	9
Age, mean ± SD (years)	58±10	60±11	60±10	(0.51)

P value, Pearson’s chi-squared test for categorical variables and t-test for continuous variables; EGFR, epidermal growth factor receptor; OR, odds ratio; CI, confidence interval; n, number of patients; SD, standard deviation.

Discussion

Epidemiological studies have consistently demonstrated that familial cancer history increases the risk of developing lung cancer (16,17), which suggests potentially heritable genetic factors in cancer development. Indeed, heritable susceptibility mutations have been identified on chromosome 6q and EGFR-T790M in familial aggregations of lung cancer patients (18,19). Asians in particular have increased susceptibility to EGFR mutations due to a combination of polymorphisms, mutations, and amplifications, and these mutations are well documented in the pathogenesis of lung cancer (20). Therefore, Asians with lung cancer may better reflect the relationship between familial cancer history and EGFR mutations.

This study demonstrated an increased prevalence of EGFR mutations in never smoking lung cancer patients with familial cancer history. These mutations contribute to lung cancer development and may explain the strong association between family history and lung cancer risk in Asian never smokers (21). The study also revealed that a sizable proportion (40%) of lung cancer patients in Taiwan harbored EGFR mutations, compared to a lower prevalence (10-15%) reported in Caucasians (22). Several implications arise from these observations.

An increased prevalence of EGFR mutations in Asians with lung cancer implies that more patients may respond to tyrosine kinase inhibitors (TKI). Overall, TKI yielded favorable clinical response in lung cancer patients with EGFR mutations (23). TKI also significantly delayed disease progression in patients with advanced cancer (24). As shown in this study, lung cancer patients with familial cancer history commonly harbored EGFR mutations and may therefore benefit from TKI therapy. For instance, patients with familial cancer history suffering from life threatening conditions such as brain metastasis should be considered for empirical TKI treatment. Of course, these patients should eventually receive EGFR mutation analysis to justify TKI therapy.

A family history specifically of lung cancer was found to increase the prevalence of EGFR mutations in never smokers with lung cancer. This finding emphasizes the role of genetic abnormalities in lung cancer development. Due to the observed familial aggregation of lung cancer patients, family members of never smoking patients with EGFR mutations may have a genetic predisposition to lung cancer development and should receive appropriate screening study. Interestingly, this study also observed a significant association between EGFR mutations and family history of non-pulmonary cancers, although the association was stronger in patients with familial lung cancer history. Family history of non-pulmonary cancers may affect the prevalence of EGFR mutations in never smokers with lung cancer because a number of different tumors such as squamous carcinomas have also been associated with these mutations (25). Of course, EGFR mutations in lung cancer and non-pulmonary cancers may occur at different genetic loci, and the role of EGFR mutations in different cancers requires further investigation.

Analysis of patients with concomitant familial cancer history and EGFR mutations showed a tendency towards more advanced disease at diagnosis. This reinforces the previous finding that patients with familial lung cancer history frequently experienced worse clinical outcome than those without family history (26). A possible explanation may lie in the oncogenic effects of EGFR mutations on lung cancer cells, including promotion of cancer proliferation, angiogenesis and metastasis (27). Since patients with positive familial cancer history were also shown to have higher prevalence of EGFR mutations, they should receive more comprehensive survey at diagnosis to identify evidence of advanced disease.

Never smokers with lung cancer were frequently women in this study, which is compatible with previous literature (28). That Asian women were less likely to smoke partly explained their increased representation in never smokers with lung cancer, but hormonal factors and cooking practice also contributed to cancer development in this population (29-31). Specifically, estrogen receptors may interact with EGFR in the pathogenesis of cancer in women (32).

The study provides novel information by examining a homogenous cohort of Taiwanese never smokers with lung cancer. Since cigarette smoking is an independent risk factor for developing lung cancer, never smokers better reflect the genetic factors associated with cancer development. Furthermore, all study subjects had complete documentation of familial cancer history and tumor genotype. Selection bias was avoided because of similar demographic characteristics between cases and controls. Recall bias was overcome by extracting data from existing medical records.

This study had several limitations. Firstly, environmental exposure was a potential confounding factor. However, since participants resided in the same municipality in central Taiwan, baseline environmental exposure such as air pollution should be comparable. Secondly, since never smoking lung cancer patients in Taiwan were more likely to be women, the study’s results may not be applicable to other populations. Thirdly, further sub-classification of familial cancer history according to degree of consanguinity would allow better understanding of its effects on the prevalence of EGFR mutations in never smoking lung cancer patients.

In conclusion, familial cancer history increased the prevalence of EGFR mutations in Taiwanese never smokers with lung cancer. Lung cancer patients with familial cancer history frequently harbored EGFR mutations and may respond favorably to TKI, and a family history of cancer can guide empirical TKI therapy in patients with life threatening disease. In essence, lung cancer in never smokers may be a distinct entity in which genetic factors play an important role.